Multi-drug resistant (MDR) organisms due to Gram negative bacilli (GNB) have become endemic in healthcare systems throughout the world. Of great concern is the recent, remarkable rise in the frequency of extremely drug resistant (XDR)-A. baumannii (AB) infections which increased >17 fold in the last decade. Further, infections with AB are surpassing those caused by Pseudomonas aeruginosa (PA), another MDR GNB, the incidence of which has remained stable across the US. Bloodstream infections caused by XDR AB have >60% mortality rates. Clearly new methods to prevent and treat AB and other MDR GNB are needed. Septicemia is a common manifestation of AB and other GNB infections which is caused by the endotoxic bacterial lipopolysaccharide (LPS) through binding to TLR4. LPS recognition by TLR4 triggers a potent and protective inflammatory cytokine immune response via the MyD88/NF-kB cascade. However, this immune response can often lead to a harmful and frenetic process associated with excess vascular leak that leads to tissue edema, organ failure, shock, and death. We found TLR4 deficient mice to be resistant to AB infection. Consistent with these findings, disruption of LPS synthesis in AB (with LpxC inhibitors affecting lipid A biosynthesis) renders the bacterium totally avirulent in mice despite the lack of in vitro activity of these inhibitors against the bacterium. Recent research by Dr. Dean Li (CSO of Navigen) demonstrated that the excess vascular leak seen in LPS-triggered septicemia is caused by the small GTPase Arf6 which falls in the MyD88/ARNO/Arf6 pathway and functions independently from the MyD88/NF-kB cascade. Arf6 induces vascular endothelium leak via intracellular internalization of VE-cadherin which leads to cell-to-cell junction disruption. We hypothesize that AB bacteremia (and other GNB infections) often fail antibiotic therapy because LPS induces excess vascular leak that results in edema and organ failure. Also, the pharmacological inhibition of Arf6 provides an opportunity to retain the beneficial inflammatory immune response, trigged by TLR4 recognizing LPS, without the added harmful vascular leak. Indeed our preliminary results showed promising activity of one of the Arf6 inhibitors in the mouse model of AB infection. We propose to build on these exciting data to delineate the role of MyD88/ARNO/Arf6 in the pathogenesis of AB infections via siRNA gene inhibition, peptide analogues and knockout mice. Further, we will determine the activity of 2nd generation improved Arf6 inhibitors in treating murine AB infections. Achieving these two goals (R21 aims) will identify 2 lead Arf6 inhibitors for detailed PK/PD and toxicology studies. Finally, we will determine the breadth of activity of these 2 inhibitors against other GNB infections including MDR PA and Klebsiella pneumonia carbapenemase (KPC) in mice (R33 aims). Conclusion of our studies will lay the foundation for further development of Arf6 inhibitors as viable new class of antibiotics versus GNB infections including GMP, GLP preclinical toxicity, and IND filing.